Compte-rendu de l’École Thématique Sciences de l’Information Géographiques Reproductibles SIGR

23/07/2021 · Oléron, 28 juin au 2 juillet 2021

# titre 1
## titre 2
*italique*
**gras**
print ("hello")
## [1] "hello"
DiagrammeR::mermaid("
graph TD;
A[point A] --> B{à pied}
A --> C{à vélo}
A --> D{à cheval}
B --> E[point B]
C --> E[point B]
D --> E[point B]
", height = 300)
help(package="spData")
library(help=spData)
help(afcon)
class(afcon)
## [1] "data.frame"
class(elev)
## [1] "RasterLayer"
## attr(,"package")
## [1] "raster"
afcon <- spData::afcon
head(afcon)
##             x        y totcon       name   id
## TS   9.556491 34.11685   1363    TUNISIA 2040
## AG   2.631669 28.16258   1421    ALGERIA 2039
## MO  -6.316814 31.88583   1861    MOROCCO 2038
## LY  18.022672 27.04399   2355      LIBYA 2041
## EG  29.774473 26.56703   5246      EGYPT 2043
## MR -10.332211 20.25840    811 MAURITANIA 2005
northland_lonlat = sf::st_transform(nz[1,], 4326) # nz est dispo via le package spData, ici on ne prend qu'une seule région
schools_nl_osm = opq(bbox = sf::st_bbox(northland_lonlat)) %>% 
  add_osm_feature(key = "amenity", value = "school") %>% 
  osmdata_sf()
schools_nl_polygons = schools_nl_osm$osm_polygons
schools_pol_nogeom <- st_drop_geometry(schools_nl_polygons)
skim(schools_pol_nogeom)
place_name = "Canberra"
et = c("amenity")
q_points = "SELECT * FROM points WHERE amenity IN ('school')"
oe_school_points = oe_get(place_name, provider = "openstreetmap_fr", query = q_points, extra_tags = et)
mapview::mapview(oe_school_points)
oe_providers()
##   available_providers          database_name number_of_zones number_of_fields
## 1           geofabrik        geofabrik_zones             430               14
## 2              bbbike           bbbike_zones             235               10
## 3    openstreetmap_fr openstreetmap_fr_zones             903                6
plot(st_geometry(openstreetmap_fr_zones))
nz_elev = rast(system.file("raster/nz_elev.tif", package = "spDataLarge"))
tm_shape(nz_elev) +
  tm_graticules(col = "white") +
  tm_raster(title = "Elevation (m asl)", 
            style = "cont",
            palette = brewer.pal(6,"YlOrBr")) +
  tm_shape(nz) +
  tm_borders(col = "darkgray", 
             lwd = 1) +
  tm_shape(nz_height) +
  tm_symbols(size = 0.2, 
             shape = 2,
             col = "black") +
  tm_add_legend(type = "symbol",
                shape = 2,
                size = 0.4,
                col = "black",
                title = "Highest peaks") +
  tm_scale_bar(breaks = c(0, 200),
               text.size = 0.8,
               lwd = 0.1) +
  tm_credits(text = "J. Nowosad, 2021") +
  tm_layout(bg.color = "lightblue",
            inner.margins = c(0, 0, 0, 0))
help(tm_graticules)
tmaptools::palette_explorer()
tmap_mode("view") # plot pour le mode statique, view pour le mode interactif

tm_shape(nz_elev) +
  tm_graticules(col = "white") +
  tm_raster(title = "Elevation (m asl)", 
            style = "cont",
            palette = brewer.pal(6,"YlOrBr")) +
  tm_shape(nz) +
  tm_borders(col = "darkgray", 
             lwd = 1) +
  tm_shape(nz_height) +
  tm_symbols(size = 0.2, 
             shape = 2,
             col = "black") +
  tm_add_legend(type = "symbol",
                shape = 2,
                size = 0.4,
                col = "black",
                title = "Highest peaks") +
  tm_scale_bar(breaks = c(0, 200),
               text.size = 0.8,
               lwd = 0.1) +
  tm_credits(text = "J. Nowosad, 2021") +
  tm_layout(main.title = "New Zealand",
            bg.color = "lightblue",
            inner.margins = c(0, 0, 0, 0))
help(tmap_save)
## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
## Variable(s) "resMco" contains positive and negative values, so midpoint is set to 0. Set midpoint = NA to show the full spectrum of the color palette.
r[r[]== -9999] <- NA
r[is.na(r)] <- 0
# importe le jeu de données exemple sur la Martinique
mtq <- mf_get_mtq()
# utilise le thème algolight
mf_theme("agolalight")
# crée la carte de base
mf_map(x = mtq)
# ajoute des symboles proportionnels colorés en fonction du revenu médian
mf_map(x = mtq, var = c("POP","MED"), type = "prop_choro")
mf_title("Population and Wealth")
srtm_path = system.file("raster/srtm.tif", package = "spDataLarge")
srtm = rast(srtm_path) # terra
srtm_r = raster(srtm_path) # raster
srtm_s = read_stars(srtm_path) # stars
plot(srtm)
srtm2 = srtm + 1000
srtm3 = srtm - global(srtm, min)[[1]]
srtm4 = srtm - global(srtm, median)[[1]]
plot(c(srtm, srtm2, srtm3, srtm4))
rcl = matrix(c(0, 1500, 1, 
               1500, 2000, 2, 
               2000, 9999, 3),
             ncol = 3, 
             byrow = TRUE)
srtm_recl = classify(srtm, rcl = rcl)
plot(srtm_recl)
srtm_focal_mean = focal(srtm, 
                   w = matrix(1, nrow = 9, ncol = 9), 
                   fun = mean)
plot(c(srtm, srtm_focal_mean))
nlcd = rast(system.file("raster/nlcd2011.tif", package = "spDataLarge"))
plot(nlcd)
crs(nlcd, describe = TRUE)
##                               name EPSG area         extent
## 1 UTM Zone 12, Northern Hemisphere <NA> <NA> NA, NA, NA, NA
crs(srtm, describe = TRUE)
##     name EPSG area         extent
## 1 WGS 84 4326 <NA> NA, NA, NA, NA
srtm_utm = project(srtm, nlcd, method = "bilinear")
srtm_zonal = zonal(srtm_utm, nlcd, na.rm = TRUE, fun = "mean")
srtm_zonal
##    nlcd2011     srtm
## 1        11 2227.060
## 2        21 1713.980
## 3        22 1642.077
## 4        23 1569.632
## 5        31 1854.069
## 6        41 2361.121
## 7        42 1867.068
## 8        43 2500.253
## 9        52 1650.966
## 10       71 1644.359
## 11       81 1284.106
## 12       82 1417.671
## 13       90 1254.168
## 14       95 1909.590
global(srtm, fun = "mean")
##          mean
## srtm 1842.548
freq(nlcd)
##       layer value  count
##  [1,]     1    11   1209
##  [2,]     1    21  14149
##  [3,]     1    22   3173
##  [4,]     1    23    195
##  [5,]     1    31 106070
##  [6,]     1    41 196044
##  [7,]     1    42 564668
##  [8,]     1    43   6825
##  [9,]     1    52 545771
## [10,]     1    71   4878
## [11,]     1    81   8460
## [12,]     1    82    268
## [13,]     1    90   6422
## [14,]     1    95     75
zion = read_sf(system.file("vector/zion.gpkg", package = "spDataLarge"))
srtm_utm_c = crop(srtm_utm, vect(zion))
srtm_utm_m = mask(srtm_utm_c, vect(zion)) # on combine généralement crop et mask
## tmap mode set to plotting
tc1 = tm_shape(srtm_utm) +
  tm_grid() + 
  tm_raster(style = "cont") +
  tm_shape(zion) +
  tm_borders(col = "black") +
  tm_layout(inner.margins = 0)
tc2 = tm_shape(srtm_utm_c) +
  tm_grid() + 
  tm_raster(style = "cont") +
  tm_shape(zion) +
  tm_borders(col = "black") +
  tm_layout(inner.margins = 0)
tc3 = tm_shape(srtm_utm_m) +
  tm_grid() + 
  tm_raster(style = "cont") +
  tm_shape(zion) +
  tm_borders(col = "black") +
  tm_layout(inner.margins = 0)
tmap_arrange(tc1, tc2, tc3)
zion_points = read_sf(system.file("vector/zion_points.gpkg", package = "spDataLarge"))
zion_extract = terra::extract(srtm, vect(zion_points))
head(zion_extract)
##   ID srtm
## 1  1 1802
## 2  2 2433
## 3  3 1886
## 4  4 1370
## 5  5 1452
## 6  6 1635
zion_points2 = cbind(zion_points, zion_extract)
plot(zion_points2['srtm'])
zion = read_sf(system.file("vector/zion.gpkg", package = "spDataLarge"))
zion = st_transform(zion, crs(srtm))
zion_srtm_values = terra::extract(srtm, vect(zion))
zion_srtm_values %>% 
  group_by(ID) %>% 
  dplyr::summarise(across(srtm, list(min = min, mean = mean, max = max)))
## # A tibble: 1 x 4
##      ID srtm_min srtm_mean srtm_max
##   <dbl>    <dbl>     <dbl>    <dbl>
## 1     1     1122     1818.     2661
zion_srtm_values2 = terra::extract(srtm, vect(zion), exact = TRUE)
zion_srtm_values2 %>% 
  group_by(ID) %>% 
  summarize(across(srtm, list(min = min, mean = mean, max = max)))
## # A tibble: 1 x 4
##      ID srtm_min srtm_mean srtm_max
##   <dbl>    <dbl>     <dbl>    <dbl>
## 1     1     1119     1817.     2661
zion_points_utm = st_transform(zion_points, crs = crs(nlcd))
raster_template = rast(ext(zion_points_utm), 
                       resolution = 4000,
                       crs = crs(zion_points_utm))
ch_raster2 = rasterize(vect(zion_points_utm), raster_template, fun = length)
tm_shape(ch_raster2) +
  tm_raster(title = "Count") +
  tm_shape(zion_points_utm) + 
  tm_dots()
library(supercells)
ortho = rast(system.file("raster/ortho.tif", package = "supercells"))
plot(ortho)
ortho_slic1 = supercells(ortho, k = 200, compactness = 10)
## Step: 21
## Initialization: Completed
## Iteration: 1/10
Iteration: 2/10
Iteration: 3/10
Iteration: 4/10
Iteration: 5/10
Iteration: 6/10
Iteration: 7/10
Iteration: 8/10
Iteration: 9/10
Iteration: 10/10
## Cleaning connectivity: Completed
plot(ortho)
plot(st_geometry(ortho_slic1), add = TRUE)

Name	schools_pol_nogeom
Number of rows	142
Number of columns	28
_______________________
Column type frequency:
character	28
________________________
Group variables	None

skim_variable	n_missing	complete_rate	min	max	n_unique
osm_id	0	1.00	8	9	142
name	18	0.87	12	38	124
MOE.authority	37	0.74	5	10	3
MOE.decile	127	0.11	1	1	6
MOE.gender	37	0.74	4	5	3
MOE.id	25	0.82	1	4	116
MOE.years	39	0.73	3	5	7
addr.city	64	0.55	5	14	36
addr.hamlet	139	0.02	6	12	3
addr.housenumber	86	0.39	1	5	45
addr.postcode	74	0.48	4	4	38
addr.street	52	0.63	8	22	76
addr.suburb	131	0.08	7	11	11
amenity	0	1.00	6	6	1
contact.facebook	141	0.01	72	72	1
denomination	135	0.05	8	21	3
email	136	0.04	15	29	6
grades	139	0.02	3	4	3
landuse	141	0.01	11	11	1
name.ja	139	0.02	6	10	3
name.mi	140	0.01	18	25	2
old_name	141	0.01	32	32	1
phone	56	0.61	11	15	86
ref.linz.address_id	132	0.07	6	7	10
religion	133	0.06	9	9	1
website	48	0.66	21	74	93
wikidata	29	0.80	8	10	113
wikipedia	140	0.01	17	19	2

Characteristic	Beta	95% CI¹	p-value
(Intercept)	9.6	9.1, 10	<0.001
log_dist_litt	-0.08	-0.12, -0.04	<0.001
log_surface_reelle_bati	0.46	0.37, 0.55	<0.001
log_surface_terrain	0.22	0.18, 0.26	<0.001
¹ CI = Confidence Interval

Compte-rendu de l’École Thématique Sciences de l’Information Géographiques Reproductibles SIGR

Compte-rendu de l’École Thématique Sciences de l’Information Géographiques Reproductibles SIGR

Présentation de l’école

Liens utiles

En bref : sélection de quelques points abordés

Bonnes pratiques en matière de reproductibilité

Créer un environnement reproductible dans R

Cartographie reproductible

Télédétetection/données rasters

Pour se documenter et résoudre les problèmes

Divers

Conférences et ateliers

Reproductibility crisis and Open Science

Produire des documents reproductibles avec R Markdown

titre 1

titre 2

Présentez vos travaux en Markdown

Introduction à git

GIS and mapping - A lecture on the GIS and mapping ecosystem of R

Git avec des petits dessins

GIS and mapping workshop

Partie 1 avec Robin Lovelace : importer des données, voir les données manquantes…

Partie 2 avec Jakub Nowosad : visualiser les données avec le package tmap

Sciences reproductibles et analyse spatiale : quelques lectures et retours d’expérience

Tenir un journal de laboratoire avec emacs et orgmode

Atelier analyse spatiale (Régression Géographiquement Pondérée ou GWR)

Analyse d’image raster (et télédétection)

Faire des cartes avec le package R mapsf

Image processing and all things raster

Map Algebra

Interactions raster-vecteur

Conversions (formats)

Analyse raster

Questions, discussions et ressources générales sur l’utilisation de R

Biblio

Publications

Autres ressources (sites web)